Exploiting Polarity Features for Developing Sentiment Analysis Tool

This paper proposes a system known as: SentiFinder (Sentiment Finder), a tool for sentiment analysis of amazon data to identifying the intensity of sentiments either positive or negative. The proposed system is based on our previous comprehensive experiments which we have been doing since more than a year. To identify a Sentiment of a comment/review, one need to analyze polarity features present in the natural language text. Different researchers have utilized different polarity features like adjectives, verbs, and adverbs. To conduct this study a comprehensive dataset has been acquired which contains 53,258 from Amazon. We extracted verbs, adverbs, and adjectives and evaluated them. It is found that adverb, adjectives, and verb combination can achieve the nest accuracy when trained on a specific settings of Random Forest Classifier and Gradient Boosting Classifier. This paper explains the lessons learned from the literature and followed by the findings and it gives an input to build a scalable system: SentiFinder.